This application is submitted in response to NIMH PA-02-027. The goal of this grant is to find a good model for schizophrenia based on gila dysfunction that can be used as a screening tool for antipsychotic drugs. An intense investigation of the role of glia in a genetic model of schizophrenia is overdue, since malfunctioning gila is expected to have a marked effect on synaptic transmission, especially glutamate transmission in the hippocampus and neorcortex. This area of study has received little attention in schizophrenia, a disease in which the primary neuropathology has yet to be identified. In light of the recently found association of the neuregulin 1 (NRG1) gene and this disorder, we propose to investigate two lines of transgenic mice in which the function of the NRG receptors erbB2, 3 and 4 has been disabled through insertion of a dominant negative erbB mutation in either oligodendrocytes or astrocytes. Phase I of this grant will concentrate on the validation of these transgenic lines of gila dysfunction as a model of schizophrenia or of schizophrenic symptoms. The endpoint measures for Phase I are based on the known phenotypes of putative animal models of schizophrenia (behavioral hyperactivity and impaired prepulse inhibition of startle) and other behavioral measures covering negative, positive and cognitive symptoms of schizophrenia. We will use pharmacology to exacerbate or rescue possible behavioral differences and will characterize basic aspects of neuroanatomy and neurotransmitter function corresponding to reported dysfunctions in schizophrenic patients. The main advantage of the cell-specific models we chose to pursue resides in that they spare NRG1-erbB signaling at neuromuscular junctions, which could confound the behavioral analysis. If successful, Phase II will follow up with a full characterization of the pharmacological sensitivity of the model(s) to typical and atypical antipsychotic drugs, and the time course of recovery of those behavioral, neurotransmitter and neuroanatomical characteristics that distinguish the model from its wild type (WT) control, as found in Phase I. Commercial Value: It is an extremely high priority for the field to develop improved animal models of schizophrenia. Such models will allow testing of antipsychotic treatments based on novel and standard chemistries, and will define the preferred developmental time and drug regime for an optimal therapeutic treatment.